25
University of Groningen Kinetics, selectivity and scale up of the Fischer-Tropsch synthesis van der Laan, Gerard Pieter IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 1999 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): van der Laan, G. P. (1999). Kinetics, selectivity and scale up of the Fischer-Tropsch synthesis Groningen: s.n. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date: 03-07-2018

University of Groningen Kinetics, selectivity and scale up of … ·  · 2016-03-05the kinetics of the Fischer-Tropsch synthesis can be divided into two classes. ... derived Langmuir-Hinshelwood-Hougen-Watson

  • Upload
    dodat

  • View
    216

  • Download
    0

Embed Size (px)

Citation preview

University of Groningen

Kinetics, selectivity and scale up of the Fischer-Tropsch synthesisvan der Laan, Gerard Pieter

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:1999

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):van der Laan, G. P. (1999). Kinetics, selectivity and scale up of the Fischer-Tropsch synthesis Groningen:s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 03-07-2018

r

sutwv0xzy�t|{0y$} ~�y�t|��v���}�{�����v0�|� ���w{����������$����${�}��|����� �¡�¢�¤£|¥�}�¦ �¨§�� © �«ª¬���®­��w¥¯�°¦¨�±�¢ª²³���|}´ªµ�"�¤§|¥

Abstract

Thekineticsof thegas-solidFischer-Tropschsynthesisover a commercialFe-Cu-K-SiO2 catalystwasstudiedin a continuousspinningbasket reactor. Experimentalcon-ditionswerevariedasfollows: reactorpressureof 0.8-4.0MPa, H2/CO feedratio of0.25-4.0,andspacevelocityof 0.5-2.010¶ 3 Nm3 kg¶ 1

cat s¶ 1 ataconstanttemperatureof523K. A numberof rateequationswerederivedonthebasisof adetailedsetof possi-ble reactionmechanismsoriginatingfrom thecarbidemechanismfor thehydrocarbonformationandtheformatemechanismfor thewatergasshift reaction,respectively. 14modelsfor theFischer-Tropschreactionrateand2 watergasshift reactionratemodelswerefitted to theexperimentalreactionrates.Bartlett’s testwasusedto reducethesetof Fischer-Tropschrateequationsto 3 models,which werestatisticallyindistinguish-able. It couldbeconcludedthat the reactionrateof theFischer-Tropschsynthesisiscontrolledby the formationof the monomerspecies(methylene)by hydrogenationof molecularlyadsorbedCO, whereasthe carbondioxide formationrate (watergasshift) is determinedby theformationof a formateintermediatespeciesfrom adsorbedCO anddissociatedhydrogen.Simulationsusingtheoptimalkinetic modelsderivedshowed goodagreementboth with experimentaldataandwith somekinetic modelsfrom literature.

145

146 CHAPTER 5

5.1 Intr oduction

TheFischer-Tropschsynthesiscanbesimplifiedasa combinationof theFT reactionandthewatergasshift (WGS)reaction.Wateris aprimaryproductof theFT reaction,andCO2 canonly be producedby the WGS reaction(RWGS · RCO2). The watergasshift reactionis a reversibleparallel-consecutivereactionwith respectto CO (seeFigure5.1).

CO + H2 H2O + -(CH2)-

CO + H2O

+CO2

RFT

RWGS

Figure5.1 Schemeof thereactionof carbonmonoxideandhydrogen.

Themajorproblemin describingFischer-Tropschreactionkineticsis thecomplex-ity of its reactionmechanismandthelargenumberof speciesinvolved. Literatureonthe kineticsof the Fischer-Tropschsynthesiscanbe divided into two classes.Moststudiesaimatcatalystimprovementandpostulateempiricalpowerlaw kineticsfor thecarbonmonoxiderates[1, 2]:

¸ RCO · k PaH2

PbCO (5.1)

andcarbondioxideformationor watergasshift reaction[3, 4]:

RCO2 · k PcH2O Pd

CO (5.2)

Relatively few studiesaim at understandingthereactionmechanisms.Someauthorsderived Langmuir-Hinshelwood- Hougen-Watson(LHHW) rate expressionsfor thereactantconsumption[5, 6]. In mostcasestheratedeterminingstepwasassumedtobetheformationof thebuilding block or monomer, methylene[7–14]. Simultaneousmodelingof theWGSandFT reactionsoniron catalystswith WGSactivity hashardlybeenreported.ZimmermanandBukur [9] andShenet al. [15] fitted kinetic expres-sionsto theirdata,but their rateexpressionsfor theWGSwerelargelyempirical.

Our objective is to develop intrinsic rate expressionsfor the CO conversiontoFischer-Tropschproductsandfor the watergasshift (WGS) reactionover a precip-

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 147

itated iron catalyston the basisof realisticmechanisms.It alsoappearedthat sev-eral existing literaturemodelscan be derived from the samelimited set of mecha-nisms[7–14]. A reactormodel will be usedto predict the reactionratesand con-versionsasa functionof experimentalconditions.Comparisonbetweenthenew rateexpressionsandavailable literaturemodelsis includedaswell. The kineticsof thegas-solidFischer-Tropschsynthesisover a commercialFe-Cu-K-SiO2 catalystwerestudiedin a continuousspinningbasket reactor(CSTR)at industrially relevant con-ditions. Productdistributionsat thesamereactionconditionsarereportedin Chapter4.

5.2 Theory

5.2.1 ActiveSiteson Precipitated Ir on Catalysts

The compositionof iron-basedcatalystschangesduring Fischer-Tropschsynthesis.ZhangandSchrader[16] concludedthat two active sitesoperatesimultaneouslyonthe surfaceof iron catalysts:Fe0/Fe-carbidesand magnetite(Fe3O4). The carbidephaseis active towardsdissociationof CO andformationof hydrocarbons,while theoxidephaseadsorbsCO associatively andproducespredominantlyoxygenatedprod-ucts. Lox et al. [17] andShroff et al. [18] concludedthat the magnetitephasehasnegligible catalyticactivity towardsFT reactionswhereascarbideformationresultsinahighFT activity.

Severalauthorsproposedthatmagnetite(Fe3O4) is themostactive phasefor theWGS reaction[4, 5, 16, 19, 20] on iron catalysts. Raoet al. [19] studiedthe ironphaseof Fe/Cu/K/SiO2 catalystsfrom thedemonstrationunit at LaPorte,Texas(Au-gust,1992)with Mossbauerspectroscopy. Thechangesof themagnetitephasecorre-spondedwith theWGSreactionactivity duringtime-on-stream.Lox etal. [17] showedthatFe3O4 coexistswith variousiron carbideson thecatalystduringsynthesisgasre-actions.It is generallyacceptedthat theWGSandFT reactionsproceedon differentactivesitesonprecipitatediron catalysts[5, 19].

5.2.2 Hydr ocarbonFormation

5.2.2.1 Elementary Reactions

Themechanismof thehydrocarbonformationduringtheFTShasbeenreviewedanddiscussedby severalauthors[1, 21–24]. Recentreviews weregivenby Hindermannetal. [25], Dry [26], Dry [27], andAdesina[28] andin Chapter2. Themostimportant

148 CHAPTER 5

growthmechanismfor thehydrocarbonformationis thesurfacecarbidemechanismbyCH2 insertion[1, 6, 29,30]. Thepresenceof adsorbedmethylenehasbeenidentifiedwith isotopic-tracertechniquesonFe/Al2O3 [31].

The formationof the methylenespecieswill be discussedin more detail. Hy-drogenreactsvia eitherthedissociative adsorbedstateor in themolecularstate[32].Dissociativeadsorptionof hydrogenproceedson two freeactivesites:

H2 ¹ 2s1 º » 2Hs1 (5.3)

s1 denotesa catalyticsitewherehydrocarbonscanbeformed. Carbonmonoxidead-sorbsassociatively onanactivesite[32]:

CO ¹ s1 º » COs1 (5.4)

AdsorbedCOcanbedissociatedin asecondstep:

COs1 ¹ s1 º » Cs1 ¹ Os1 (5.5)

Surfacecarbonreactswith adsorbeddissociatedhydrogen,

Cs1 ¹ Hs1 º » CHs1 ¹ s1 (5.6)

CHs1 ¹ Hs1 º » CH2s1 ¹ s1 (5.7)

or with molecularhydrogen,

Cs1 ¹ H2 º » CH2s1 (5.8)

Oxygenis removed irreversibly andrapidly from the surfaceby consecutive hydro-genationreactions[24, 33,34],

Os1 ¹ Hs1 º HOs1 ¹ s1 (5.9)

HOs1 ¹ Hs1 º H2Os1 ¹ s1 (5.10)

H2Os1 º » H2O ¹ s1 (5.11)

or with molecularhydrogenaccordingto anEley-Ridealmechanism[8, 24,33],

Os1 ¹ H2 º H2Os1 (5.12)

H2Os1 º » H2O ¹ s1 (5.13)

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 149

Anotherpossiblemechanismstartswith molecularlyadsorbedcarbonmonoxideandsuccessivehydrogenassisteddissociationwith dissociatedhydrogen[6, 8],

COs1 ¹ Hs1 º » HCOs1 ¹ s1 (5.14)

HCOs1 ¹ Hs1 º » Cs1 ¹ H2Os1 (5.15)

or molecularhydrogen,

COs1 ¹ H2 º » HCOHs1 ¹ s1 (5.16)

HCOHs1 ¹ H2 º » CH2s1 ¹ H2O (5.17)

Basedon theseelementaryreactions,we definedfour differentpossiblemecha-nisms.SeeTable5.1 for theconventionsandstateof thereactantsin theelementaryformationreactionsof methylene.Thecompletesetof elementaryreactionsfor eachmodelis givenin Table5.2.

Table5.1 Thevariouskineticmodelsconsidered,togetherwith thepresenceof thereactantsin theratedeterminingstep.Model CO H2

FT-I Dissociative DissociativeFT-II Dissociative MolecularFT-III Associative DissociativeFT-IV Associative Molecular

5.2.2.2 Kinetic RateEquations

In orderto deriverateequations,weusedtheLangmuir-Hinshelwood-Hougen-Watsonapproach,see,for example,Graafet al. [35]. For eachmodel,the possibleratede-terminingstepswere identified,while all otherstepswereassumedto be at quasi-equilibrium. The following assumptions,all basedon literature,weretaken into ac-count:

1. Reactionpathfor the CO consumptionto the monomermethylene,CH2, con-tainsoneirreversibleratedeterminingstep,in analogywith Ref. [35].

2. Steadystateconcentrationsof all intermediateson thecatalystsurface[35, 36].

150 CHAPTER 5

Table5.2 Elementaryreactionsfor FT synthesis.Model Number Elementaryreaction

FT-I 1 CO+ s1 º » COs1

2 COs1 + s1 º » Cs1 + Os1

3 Cs1 + Hs1 º » CHs1 + s1

4 CHs1 + Hs1 º » CH2s1+ s1

5 Os1 + Hs1 º HOs1 + s1

6 HOs1 + Hs1 º H2Os1 + s1

7 H2O + s1 º » H2Os1

8 H2 + 2s1 º » 2Hs1

FT-II 1 CO+ s1 º » COs1

2 COs1 + s1 º » Cs1 + Os1

3 Cs1 + H2 º » CH2s1

4 Os1 + H2 º H2Os1

5 H2O + s1 º » H2Os1

FT-III 1 CO+ s1 º » COs1

2 COs1 + Hs1 º » HCOs1 + s1

3 HCOs1 + Hs1 º » Cs1 + H2Os1

4 Cs1 + Hs1 º » CHs1 + s1

5 CHs1 + Hs1 º » CH2s1 + s1

6 H2 + 2s1 º » 2Hs1

7 H2O + s1 º » H2Os1

FT-IV 1 CO+ s1 º » COs1

2 COs1 + H2 º » H2COs1

3 H2COs1 + H2 º » CH2s1 + H2O

4 H2O + s1 º » H2Os1

1Equilibriumconstant,e.g.reactionstepFT-I1: K1 ¼ ½ COs1

PCO ½ s1

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 151

Table5.3 ReactionrateexpressionsconsideredfortheFischer-Tropschsynthesis,RFT (mol kg¶ 1

cat s¶ 1).

Model Kinetic equation

FT-I3kP1¾ 2

CO P1¾ 2H2

1 ¹ aP1¾ 2CO ¹ bPH2O

2

FT-I4kP1¾ 2

CO P3¾ 4H2

1 ¹ aP1¾ 2CO P ¶ 1¾ 4

H2 ¹ bPH2O

2

FT-II3kP1¾ 2

CO PH2

1 ¹ aP1¾ 2CO ¹ bPH2O

FT-III2kPCO P1¾ 2

H2

1 ¹ aPCO ¹ bPH2O2

FT-III3kPCO PH2

1 ¹ aPCO ¹ bPH2O2

FT-IV2kPCO PH2

1 ¹ aPCO ¹ bPH2O

FT-IV3kPCO P2

H2

1 ¹ aPCO ¹ bPH2O

3. Catalystsitesof type 1 areactive towardshydrocarbonformation,which areuniformandhomogeneouslydistributed[35, 36].

4. Initial adsorptionof hydrogenandcarbonmonoxideis in quasi-equilibriumwiththegasphaseconcentrations[24].

5. Wateris removedirreversiblyafterCOdissociation[24, 33,37].

6. CO is adsorbedmore strongly than H2 on iron catalysts,resultingin a highsurfaceconcentrationof COor dissociatedCOrelative to H2 [21, 38].

7. H2O adsorbsstronglyandmayinhibit theFT reactionrate[9].

With theseassumptions,7 differentkinetic modelsremainpossible. Thesearesummarizedin Table5.3. Thedevelopmentof thekineticequationswill beillustratedfor modelFT-II3. The modelcodesrefer to the setof elementaryreactionsandtheelementaryreactionnotatequilibrium(thatis theratedeterminingstep,soin thiscase

152 CHAPTER 5

Table5.4 Parametersfor theFT kineticmodels.Model k (x) a (x) b

(mol kg¶ 1 s¶ 1 MPax) (MPax) (MPa¶ 1)FT-I3 ¿ k3k5K1K2K8 À 1¾ 2 (-1) ¿ K1K2k5Á k3 À 1¾ 2 (-1/2) K7

FT-I4 ¿ k4k5K1K2K3 À 1¾ 2K 3¾ 48 (-5/4) ¿ K1K2K3K 1¾ 2

8 k5Á k4 À 1¾ 2 (-1/4) K7

FT-II3 ¿ k3k4K1K2 À 1¾ 2 (-3/2) ¿ k4K1K2 Á k3 À 1¾ 2 (-1/2) K5

FT-III2 k2K1K 1¾ 26 (-3/2) K1 (-1) K7

FT-III3 k3K1K2K6 (-2) K1 (-1) K7

FT-IV2 k2K1 (-2) K1 (-1) K4

FT-IV3 k3K1K2 (-3) K1 (-1) K4

reaction3). Thesetof elementaryreactionsfor modelFT-II3 is shown in Table5.2.Thereactionrateof theratedeterminingstepis:

RFT-II3 · k3Â Cs1 PH2 · k4 Â Os1 PH2 (mol kg¶ 1cat s¶ 1) (5.18)

Thesurfacefractionof carboncanbecalculatedfrom thesitebalance,theprecedingreactionstepswhichareatquasi-equilibriumandthereactionratefor waterformation:

K1 · Â COs1

PCO Â s1 Ã K2 · Â Cs1 Â Os1Â COs1 Â s1

(5.19)

 Cs1 · k4

k3Â Os1 · K1K2k4

k3

1¾ 2P1¾ 2

CO Â s1 (5.20)

Fromassumptions6 and7 it follows thatonly surfacecarbonandwateroccupy asignificantfractionof thetotalnumberof sites,thesitebalancebecomes:

 s1 ¹Ä Cs1 ¹Ä H2Os1 · 1 (5.21)

Substitutionof thesurfacefractionof carbonin eq5.18:

RFT-II3 · Å k3k4K1K2 Æ 1¾ 2P1¾ 2CO PH2

1 ¹ Å K1K2k4Ç k3 Æ 1¾ 2P1¾ 2CO ¹ K5PH2O

· kP1¾ 2CO PH2

1 ¹ aP1¾ 2CO ¹ bPH2O

(5.22)

Table5.3summarizesthefinal form of thevariousrateexpressionsfor the7 pos-siblekinetic modelsconsidered,whereasTable5.4 shows thekinetic andadsorptionparametersfor thedifferentkineticmodels.It canbeseenthatthepressuredependencyof COandH2 in thenumeratorrangesfrom 1/2to 1,and1/2to 2, respectively. Thede-nominatoris quadraticin caseof adualsiteelementaryreaction,in contrastto asingle

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 153

siteratedeterminingstep.Thedenominatorconsistsof theindividualcontributionsofsignificantlyabundantspecieson thecatalystsurface.

Theconcentrationof freesites s1 is determinedfrom asitebalance.It is assumedthatthetotalnumberof sitesis constant:

 s1 ¹ n

i È 1Â i s1 · 1 (5.23)

where  s1 is the fraction free sitesand  i s1 are the surfacefractionsoccupiedwithadsorbedspeciessuchas carbon,carbonmonoxide,hydrogen,alkyl chains,water,carbondioxide,andsoforth Theadditionof severalinhibition termsin thedenomina-tor cannot be justifiedstatisticallydueto a high degreeof covarianceor correlation[39, 40]. The derived kinetic expressionshave a maximumof two inhibition terms:onetermfor COor acarbidicspecies(  Cs1) andtheotherfor H2O inhibition.

5.2.2.3 Literatur eModels

Reviewsof kineticequationsfor iron-basedcatalystswerepublishedby Huff andSat-terfield [8], ZimmermanandBukur [9], andVan der LaanandBeenackers[45], en-closedin slightly revisedform asChapter2. Kinetic studiesof theFTSon iron andcobaltcatalystsaresummarizedin Table5.5.Thecorrespondingoperatingconditionsaregivenin Chapter2 (Table2.7).

It canbeshown thatall theseliteraturemodelscanbederivedfromthesetof mech-anismsconsideredin this studyandwhich aresummarizedin Table5.2. Appropriateassumptionsfor theinhibitor effectsin thesitebalanceof thekinetic rateexpressionsin table5.3 resultin similar mathematicalexpressions.Themechanisticimplicationsof theavailableFT kineticmodelsaresummarizedin Table5.5.

5.2.3 Water GasShift Reaction

5.2.3.1 ReactionMechanism

Several mechanismsfor the watergasshift reactionwereproposedin the literature.Singlestudiesof the watergasshift reactionover supportedmetalssuggestthe ap-pearanceof formatespecies[4, 5, 20,35]. Theformatespeciescanbeformedby thereactionbetweeneitherahydroxyspeciesor waterandcarbonmonoxideeitherin thegasphaseor in the adsorbedstate. The hydroxy intermediatecanbe formedby thedecompositionof water. Theformateintermediatecanbereducedto eitheradsorbedor gaseouscarbondioxide (seeTable5.6). RethwischandDumesic[20] studiedthewater gasshift reactionon several supportedand unsupportediron oxide and zinc

154 CHAPTER 5

Table5.5 Reactionrateequationsoverallsynthesisgasconsumptionrate.SeeTable2.7forexperimentalconditions,reactortypeandcatalystapplied.

Kinetic expression References Mechanisticimplications

(a) kPH2 [9, 22,41] FT-II3 (b=0,aPCO É 1)FT-IV2 (b=0,aPCO É 1)

(b) kPaH2

PbCO [3] -

(c)kPH2 PCO

PCO ¹ K PH2O[7, 9, 10,15,22] FT-IV2 (aPCO andbPH2OÉ 1)

(d)kP2

H2PCO

PCO PH2 ¹ K PH2O[8, 15,42,43] FT-II3 (waterformationis

reversible)

(e)kP2

H2PCO

1 ¹ aPCO P2H2

[22] -

(f)kPH2 PCO

PCO ¹ K PCO2

[9, 11,12,43] FT-IV2 (aPCO É 1 andCO2 inhibition)

(g)kPH2 PCO

PCO ¹ K1PH2O ¹ K2PCO2

[9, 11,12] FT-IV2 (aPCO É 1, CO2

andH2O inhibition)

(h)kP1¾ 2

CO P1¾ 2H2

1 ¹ K1P1¾ 2CO ¹ K2P1¾ 2

H2

2 [14] FT-I3 (H2 inhibition,bPH2O · 0)

(i)kPCO P1¾ 2

H2

1 ¹ K1PCO ¹ K2P1¾ 2H2

2 [6] FT-III2 (H2 inhibition,bPH2O · 0)

(j)kPCO PH2

Å 1 ¹ K PCO Æ 2 [13, 40,44] FT-III3 (bPH2O · 0)

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 155

Table5.6 Elementaryreactionsfor thewatergasshift reaction.Model Number Elementaryreaction

WGS-I 1 CO+ s2 º » COs2

2 CO2 + s2 º » CO2s2

3 H2O + s2 º » H2Os2

4 H2 + 2s2 º » 2Hs2

5 COs2 + H2Os2 º » HCOOs2 + Hs2

6 HCOOs2 + s2 º » Hs2+ CO2s2

WGS-II 1 CO+ s2 º » COs2

2 CO2 + s2 º » CO2s2

3 H2O + s2 º » H2Os2

4 H2Os2 + s2 º » OHs2 + Hs2

5 H2 + 2s2 º » 2Hs2

6 COs2 + OHs2 º » HCOOs2 + s2

7 HCOOs2 + s2 º » Hs2+ CO2s2

oxide catalysts.They suggestedthat the WGS reactionover unsupportedmagnetiteproceedsvia a directoxidationmechanism,while all supportediron catalystsoperatevia a mechanismwith formatespeciesdueto limited changeof oxidationstateof theiron cations.

5.2.3.2 Kinetic Expressions

Severalassumptionsweremadein orderto derive theLHHW rateexpressions:

Ê Steadystatefor theadsorbedspecies.

Ê Oneratedeterminingstepin thesequenceof elementaryreactionsoverthecom-pleterangeof experimentalconditions.

Ê Surfaceconcentrationsof intermediatespeciesarenegligible [35].

Ê Activesitesfor theWGS(type2) aredifferentthanthesitesfor thehydrocarbonforming reactions(type1) [5].

156 CHAPTER 5

Ê Ratedeterminingstepis a dual-siteelementaryreactionbetweentwo adsorbedspecies[5].

Ê Adsorptionof reactantsanddesorptionof productsareatequilibrium.

With the mentionedassumptionstwo ratedeterminingstepsarepossible. First, theratedeterminingstepis (WGS-II6):

COs2 ¹ OHs2 º » HCOOs2 ¹ s2 (5.24)

Thehydroxylspeciesis formedby dissociationof water:

H2O ¹ 2s2 º » OHs2 ¹ Hs2 (5.25)

Second,thereactionbetweenadsorbedwaterandcarbonmonoxide(WGS-I5)canberatedetermining:

COs2 ¹ H2Os2 º » HCOOs2 ¹ Hs2 (5.26)

On basisof the formatemechanismandthe mentionedassumptions,two kineticrateequationsweredeveloped.Theexpressionsaregivenin Table5.7.Theadsorptionof H2 andCO2 areassumedto benegligible relative to CO andH2O [5, 9, 38]. Thus,themassbalanceof thecatalyticsites,s2, is:

 s2 ¹Ä H2Os2 ¹Ä COs2 · 1 (5.27)

Derivation of otherkinetic expressionsbasedon adsorptionof morecomponentsispossiblefrom theaboveequations.SincetheWGSreactionis anequilibriumreaction,thereversereactionhasto betakeninto account.For the temperaturedependency ofthe equilibrium constantof the WGS reaction,K P, the following relationwasused(Graafet al. [46]):

log K P · logPCO2 PH2

PH2O PCO eq· 2073

T¸ 2 Ë 029 (5.28)

Kinetic studiesof theWGSreactionunderFT conditionson iron-basedcatalystsaresummarizedin Chapter2 (Table2.8).

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 157

Table5.7 Rateexpressionsconsideredfor thewatergasshift reaction,RCO2 (mol kg¶ 1

cat s¶ 1).

Model Kinetic equation Sitebalance

WGS-I5kÌ PCO PH2O

¸ PCO2 PH2 Ç K P

1 ¹ K1PCO ¹ K3PH2O2 s2, COs2, H2Os2

kÌ = k5K1K3 (mol kg¶ 1 s¶ 1 MPa¶ 2)

WGS-II6kÌ PCO PH2O Ç P1¾ 2

H2¸ PCO2 P1¾ 2

H2 Ç K P

1 ¹ K1PCO ¹ K3PH2O2 s2, COs2, H2Os2

kÌ = k5K1K3K4K ¶ 1¾ 25 (mol kg¶ 1 s¶ 1 MPa¶ 1Í 5)

5.3 Experimental

The kineticsof both the Fischer-Tropschsynthesisand the watergasshift reactionover a commercialprecipitatediron catalyst(RuhrchemieLP33/81)wereunraveledby relevantexperimentsin a SpinningBasket Reactor(SBR).For a detaileddescrip-tion of theexperimentalset-up,thecatalystapplied,theanalyticandtheexperimentalprocedures,seeChapter3.

Thebasketswereloadedwith 2.34g of catalyst,with particlediametersbetween0.125and0.160mm. Thecatalystwaspretreatedwith a hydrogenflow rateof 8.3310¶ 4 Nm3 kg¶ 1

cat s¶ 1 accordingto Bukur et al. [47]. Thereactortemperaturewaslin-early increasedfrom 293 to 553K by 0.017K /s. The reactortemperaturewaskeptat 553 K for 24 hrs at atmosphericpressure.After catalystreduction,synthesisgaswas fed to the reactorwhich at standardconditionsoperatedat 523 K, 1.50 MPa,(H2/CO)f eed=2 andaspacevelocityof 1.5110¶ 3 Nm3 kg¶ 1

cat s¶ 1.

Checkingthecriteriaof WeiszandPrater[48] for thereactantsCOandH2 showedthatnointraparticlediffusionlimitationsoccurredatrelevantexperimentalconditions,evennot at thehighestconversionrates.Here,it wasconservatively assumedthatthecatalystporeswerefilled with long-chainhydrocarbonwaxes.

24 kinetic experimentswerecarriedout in theSBRwith theRuhrchemieprecipi-tatediron catalyst.Theexperimentalconditionswerevariedasfollows: P= 0.8 - 4.0MPa,H2/COfeedratio=0.25- 4.0,and Î i nÏ$Ð 0Ç W= 0.510¶ 3 - 2.010¶ 3 Nm3 kg¶ 1

cat s¶ 1 ata temperatureof 523K. At regularintervals,thestandardexperimentwasrepeatedto

158 CHAPTER 5

determinepossibledeactivationeffectsof thecatalyst.A summaryof theexperimentalresultsandoperatingconditionsis givenin AppendixA.

5.4 Resultsand Discussion

After an initial periodof 100hrs,a steadystatewasmoreor lessobtained.Thecat-alyst activity, reactionrateto hydrocarbonproducts(RFT ) andthe rateof the watergasshift (RWGS) did not changemuchover 1200hrs time on stream(seeChapter3;Figure3.10a).Thereactionrateswerenotcorrectedfor catalystagingdueto thesmalleffectof timeonstreamon thecatalystactivity.

The preliminary screeningof the Fischer-Tropschkinetic expressionswas per-formedwith a maximumof two adsorbedspeciesin the site balance.Every kineticmodelwasoptimizedwith two differentmathematicalformsof thesitebalance:

 s1 ¹ Å Â Cs1 or  COs1 Æ ¹Ä H2Os1 · 1 (5.29)

Å Â Cs1 or  COs1 Æ ¹Ä H2Os1 · 1 (5.30)

For modelsbasedon thecarbidemechanism(FT-I, FT-II), thecarbidicspeciesis sur-facecarbon Cs1, formedby dissociationof CO.ModelsFT-III andFT-IV arebasedonassociative adsorbedCO species COs1. The7 kinetic equationswereoptimizedwitha non-linearoptimizationroutineusingboth eqs5.29 and5.30 for the site balance.Contributionsof speciesin the site balancewereeliminatedif the fitted adsorptioncoefficientswerenot significantlydifferentfrom zeroor hada significantlynegativevalue.Table5.8shows theresultsof thekineticmodelswith therelativevarianceandtheirranking.Fourmodelsareableto describetheexperimentalFT reactionrateswitha relativevariancelessthan35% andamaximumof threeoptimizedparameters.

Bartlett’s testwasappliedto investigatewhetherthedifferencesin accuracy of thevariousmodelswerestatisticallysignificant[49, 50]. This test comparesa criticalcalculatedÑ 2

c value(for details,seeChapter3 or Jonker et al. [49]) with a tabulatedÑ 2t value[51]. If Ñ 2

c exceedsthetabulatedvalue,themodelwith thelargestdeviationwasrejectedand Ñ 2

c wasrecalculated.Modelsweresubsequentlyrejected,until Ñ 2c

wasbelow thetabulatedvalue.Table5.9comparesÑ 2c with thetabulatedÑ 2

t valueforH ¸ 1 degreesof freedom.Thetableshows thatthebestfivemodels(H · 5) passedthetestandarestatisticallyindistinguishable.Thebestfivemodelsare,in succeedingorder: FT-IV2 (eq 5.29), FT-III2, FT-III3, FT-IV2 (eq 5.30), FT-II3. Model FT-II3wasrejectedfrom list of bestmodels,becausetheoptimizedparametersof thismodelwereunrealisticandthemodeljust passedtheBartlett’s testdueto the large relative

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 159

Table5.8 FT kineticmodelscreening.

Model Sitebalance(eq) sr el (%) Rank Remarks

FT-I3 5.30 63.7 6FT-I4 5.30 65.9 9FT-II3 5.30 45.2 5FT-III2 5.29 30.0 2FT-III2 5.30 64.4 8FT-III3 5.29 30.9 3FT-III3 5.30 63.8 7FT-IV2 5.29 29.6 1FT-IV2 5.30 32.4 4FT-IV3 5.29,5.30 - - a Ò 0

FT-I3, FT-I4,FT-II3 with sitebalance5.29resultsin a Ó 0

variance. The site balancesof modelsFT-IV2 (eq 5.29) andFT-IV2 (eq 5.30) varyslightly only. Consequently, themodelwith thehighestrelativevariancewasrejected,i.e. FT-IV2 (eq5.30).

Four experimentswerefoundto beoutliersin thethreeremainingmodels(Runs:5, 17,20,23). Theremainingmodelswerefittedagainwith thereduceddatasetof 20experimentalreactionrates.Table5.10givestheoptimizedvaluesof theparametersinthesethreemodels:FT-III2, FT-III3, FT-IV2. Thethreeremainingkineticexpressions(FT-III2, FT-III3, andFT-IV2) areall basedonthecombinedenol/carbidemechanism.Themathematicalform of theequationsis verysimilar, indicatinga difficult discrim-inationprocedure.Figure5.2comparestheexperimentalandcalculatedreactionratesof thesethreemodels.

Kinetic expressionFT-IV2 is similar to severalliteraturemodels[7, 9, 10,22] foriron catalysts(seeTable5.5). In this model,theratedeterminingstepis a singlesitereactionbetweenundissociatedadsorbedCO andgaseousH2. However, theliteraturemodelsweredevelopedfrom experimentsin slurryphaseor packedbedreactors.Themajor differencebetweentheseliteraturemodelsandoptimizedmodelFT-IV2 is asignificantnumberof free sitesin the latter model. In our experiments,the catalystparticlesarelocatedin spinningbasketswith a small amountof high-boilinghydro-carbonspresentin thecatalystpores.

Kinetic expressionsFT-III2 andFT-III3 arealsoableto describeour experimentsaccurately. Thesemodelsarealsodevelopedfrom thecombinedenol/carbidemecha-

160 CHAPTER 5

R FT exp Ô . (mol kg -1 s Õ -1 )

Ö0.000 ×

0.002 ×

0.004 ×

0.006 ×

0.008 ×

0.010 ×

0.012 ×

R FT

mod

(m

ol k

g -1 s -1

)

0.000 ×0.002 ×0.004 ×0.006 ×0.008 ×0.010 ×0.012 ×

FT-III2 FT-III3 FT-IV2

-25 %

+25 %

cat

cat ØFigure5.2 Parity graphof experimentalandoptimizedFischer-Tropschreactionrates.

R WGS exp Ù (mol kg -1 s Ú -1 )

Û0.000 Ü

0.001 Ü

0.002 Ü

0.003 Ü

0.004 Ü

0.005 Ü

0.006 Ü

R W

GS

mod

(m

ol k

g -1 s -1

)

0.000 Ü0.001 Ü0.002 Ü0.003 Ü0.004 Ü0.005 Ü0.006 Ü

WGS-I5 WGS-II6

+ 25% Ý

- 25%

cat Þ

cat

Figure5.3 Parity graphof experimentalandoptimizedWGSreactionrates.

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 161

Table5.9 Bartlett’s testfor FT models1.H2 Ñ 2

c Ñ 2t

9 42Ë 2 15Ë 58 38Ë 2 14Ë 17 32Ë 6 12Ë 66 23Ë 6 11Ë 15 6 Ë 03 9 Ë 494 0 Ë 206 7 Ë 813 0 Ë 040 5 Ë 992 0 Ë 0037 3 Ë 841 ß 2

c : critical ß 2 accordingto Bartlett’s test[49]; ß 2t : tabulatedß 2 [51]

2 H : numberof modelsunderconsideration

nism: ratedeterminingstepsarethedualsitesurfacereactionbetweenundissociatedadsorbedCO anddissociatedH2 (FT-III2) andbetweenadsorbedformyl anddisso-ciatedH2 (FT-III3). Model FT-III2 is similar to the optimal equationof SarupandWojciechowski [14] for a precipitatedcobaltcatalystin a Berty reactor, whereasFT-III3 wasfoundto bethebestmodelby YatesandSatterfield[40] on cobaltmeasuredin a slurry reactor. The kinetic modelof SarupandWojciechowski [14] wasdevel-opedwith theassumptionthatthesitebalanceconsistsof freesites,adsorbedCOanddissociatedH2, while YatesandSatterfield[40] only includedCOinhibition.

The WGSreactionratewasoptimizedwith thekinetic expressionsin Table5.7.Dueto ahighdegreeof similarity betweenequationsWGS-I5andWGS-II6, therela-tivevariancesarealmostequal,25.0% and23.0%, respectively. TheBartlett’s testisunableto discriminatebetweenthesemodels.A parity plot betweentheexperimentalandmodelvaluesof the WGS reactionratesis shown in Figure5.3. ReactionrateexpressionWGS-II6 is similar to the optimal modelof Lox andFroment[5]. Bothmodelsassumethattherateof theWGSreactionis determinedby thereactionof ad-sorbedcarbonmonoxideandhydroxyl towardsa formateintermediate.Our modelassumesadsorptionof CO andwaterto bedominantin thesitebalance,whereasLoxand Froment[5] includedinhibition of hydroxyl speciesonly. The correspondingmodelparametersarealsogivenin Table5.10.

Both theexperimentalandthecalculatedratesof theFischer-Tropschandthewa-ter gasshift reactionarecomparedin Figures5.4- 5.5 at variousexperimentalcon-ditions. The calculatedratesstemfrom modelsFT-III2 andWGS-II6 with the input

162 CHAPTER 5

Table5.10 Finalestimatesfor theparametersof theFT andWGSkineticmodels.

Parameter Dimension Estimate

WGS-I5(sr el 21.0%)kÌ mol kg¶ 1 s¶ 1 MPa¶ 2 1.77 à 0.04K1 MPa¶ 1 2.10 à 0.04K3 MPa¶ 1 24.19à 3.14

WGS-II6 (sr el 21.5%)kÌ mol kg¶ 1 s¶ 1 MPa¶ 1Í 5 1.13 à 0.01K1 MPa¶ 1 2.78 à 0.04K3 MPa¶ 1 12.27à 0.94

FT-III2 (sr el 23.7%)k mol kg¶ 1 s¶ 1 MPa¶ 1Í 5 0.0488à 0.0049a MPa¶ 1 0.563à 0.094b MPa¶ 1 4.05 à 0.77

FT-III3 (sr el 22.4%)k mol kg¶ 1 s¶ 1 MPa¶ 2 0.0556à 0.0056a MPa¶ 1 0.125à 0.069b MPa¶ 1 7.00 à 0.87

FT-IV2 (sr el 22.7%)k mol kg¶ 1 s¶ 1 MPa¶ 2 0.0779à 0.0157a MPa¶ 1 0.536à 0.333b MPa¶ 1 32.27à 8.69

valuesof theH2/CO feedratioandtheflow rate, Î i nÏ"Ð 0 Ç W basedon stoichiometryandmassbalancesof thecomponentspresent(CO, CO2, H2, H2O). Theproductcompo-sition wasdeterminedfrom gaschromatographicanalysisof thegasandfrom liquidhydrocarbonproductsamples.Thevaluesof n andm weredeterminedfrom theprod-uct composition.In this study, n variedbetween2.86-5.04.The ratio of mÇ n variedbetween2.14-2.42.Theeffluentflow ratewasestimatedwith theaveragecontractionfactorcalculatedat mÇ n of 2.30. Sincethevariationof mÇ n with processconditionsis minor, thisassumptionseemsjustified.

Theeffect of theflow rateon theoverall rateandratesof thewatergasshift andFischer-Tropschreactionis demonstratedin Figure5.4a. As expected,the reactionratesincreasewith increasingspacevelocity. Thereis goodagreementbetweenthemodelcalculationsandtheexperimentalvalues.Overall conversionof synthesisgas,

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 163

R (

mol

kg-1

s-1)

cat

XC

O+

H2

Figure 5.4 Reactionratesfor theWGSandFT andtotal conversionof CO andH2 (a) andoverall conversionof synthesisgas(XCO á H2) versusspacevelocity (b). Symbolsareexper-imentalvalues.Linesaremodelpredictions(FT-III2 andWGS-II6).

164 CHAPTER 5

PCO (MPa) (feed)

0.0 0.5 1.0 1.50.000

0.005

0.010

0.015

0.020

0.025

0.030

-RCO+Hâ

RFT

RWGS

PH2ä (MPa) (feed)

0.5 1.0 1.5 2.0 2.5 3.00.000

0.005

0.010

0.015

0.020

0.025

0.030

RWGSåRFT

-RCO+Hâ

R(m

ol k

g-1

s-1

)ca

ta.æ b.

R(m

ol k

g-1

s-1

)ca

t

Figure 5.5 Reactionratesfor theWGSandFT andtotal conversionof CO andH2 versusreactantfeedpressures.Symbolsareexperimentalvalues.Linesaremodelpredictions(FT-III2 andWGS-II6). a: FeedpressurePCO= 0.8 MPa, T= 523 K, ç i nÏ Á W= 1.0 10¶ 3 kg¶ 1

cats¶ 1; b: FeedpressurePH2= 0.8MPa,T= 523K, ç i nÏ Á W= 1.010¶ 3 kg¶ 1

cat s¶ 1.

XCO á H2, at the sameconditionsis accuratelypredictedwith the optimizedkineticexpressionsanda CSTRreactormodel(seeFigure5.4b).

Theeffectof theindividual reactantpressures(PCO andPH2) in thefeedstreamisshown in Figures5.5a-b. Themodelsappearto predictthetrendsof varyingreactantpressuressatisfactory. Boththewatergasshift aswell astheFischer-Tropschreactionrateincreasewith increasingfeedpressureof CO (Figure5.5a). The waterpressuredecreaseswith increasingCO pressurecausingan increaseof the Fischer-Tropschreactionrate. Figure5.5b shows that the spacetime yield and the Fischer-Tropschreactionrateincreaseslightly andthendecreasemonotonically. This is causedby anincreaseof thehydrogenandwaterpressurein thereactorwith increasingfeedpres-sureof hydrogen.Wateris a stronginhibitor on thecatalystandreducesthereactionratesof thehydrocarbon-formingreactions.

5.5 Conclusions

Experimentsfor thekineticsof thehydrocarbonformationandwatergasshift reactionoveraniron catalystwereobtainedovera wide rangeof industriallyrelevantreactionconditions.A numberof rateequationswerederivedon thebasisof a detailedsetofpossiblereactionmechanisms.Thefollowing conclusionscanbemade:

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 165

1. Two differentsitesarepresenton iron catalysts.The iron carbidesareactivetowardshydrocarbonforming reactions,whereasmagnetite(Fe3O4) seemstobethemostactivesitefor thewatergasshift reaction.

2. Thereactionrateof theFischer-Tropschsynthesisis determinedby theforma-tion of themonomerspecies(methylene).Thebestmodelsassumethattheratedeterminingstepproceedsvia hydrogenationof associativeadsorbedCO.

3. Carbondioxideis formedby thewatergasshift reaction.Theratedeterminingstepis theformationof a formateintermediatespecies.

Simulationsusingthekineticmodelsderivedshow goodagreementwith bothex-perimentaldataandwith somekineticmodelsfrom literature.

References

[1] Dry, M.E., TheFischer-Tropschsynthesis,in J.R.Anderson;M. Boudart,eds.,Catalysis-Scienceandtechnology, vol. 1, Springer-Verlag,New York, 1981pp.160–255.

[2] Ribeiro,F.H.; Schachvon Wittenau,A.E.; Bartholemew, C.H.; Somorjai,G.A.,Reproducibilityof turnover ratesin heterogeneousmetalcatalysis:compilationof dataandguidelinesfor dataanalysis,Catal.Rev.-Sci.Eng. 1997, 39, 49–76.

[3] Bub,G.;Baerns,M., Predictionof theperformanceof catalyticfixedbedreactorsfor Fischer-Tropschsynthesis,Chem.Eng. Sci.1980, 35, 348–355.

[4] Newsome,D.S., The water-gasshift reaction,Catal. Rev.-Sci.Eng. 1980, 21,275–318.

[5] Lox, E.S.;Froment,G.F., Kineticsof theFischer-Tropschreactionon a precip-itatedpromotediron catalyst.2. Kinetic modeling,Ind. Eng. Chem.Res.1993,32, 71–82.

[6] Wojciechowski,B.W., Thekineticsof theFischerTropschsynthesis,Catal.Rev.-

Sci.Eng. 1988, 30, 629–702.[7] Dry, M.E., Advancesin Fischer-Tropschchemistry, Ind. Eng. Chem.Prod.Res.

Dev. 1976, 15, 282–286.[8] Huff, Jr., G.A.; Satterfield,C.N., Intrinsic kineticsof the Fischer-Tropschsyn-

thesisona reducedfused-magnetitecatalyst,Ind. Eng. Chem.ProcessDes.Dev.

1984, 23, 696–705.[9] Zimmerman,W.H.; Bukur, D.B., Reactionkineticsover iron catalystsusedfor

theFischer-Tropschsynthesis,Can.J. Chem.Eng. 1990, 68, 292–301.

166 CHAPTER 5

[10] Atwood,H.E.;Bennett,C.O.,Kineticsof theFischer-Tropschreactionover iron,Ind. Eng. Chem.ProcessDes.Dev. 1979, 18, 163–170.

[11] Nettelhoff, H.; Kokuun,R.; Ledakowicz, S.; Deckwer, W.-D., Studieson thekineticsof Fischer-Tropschsynthesisin slurry phase,Ger. Chem.Eng. 1985, 8,177–185.

[12] Ledakowicz, S.; Nettelhoff, H.; Kokuun,R.; Deckwer, W.-D., Kinetics of theFischer-Tropschsynthesisin theslurryphaseona potassium-promotediron cat-alyst,Top.Catal.1985, 24, 1043–1049.

[13] Dixit, R.S.;Tavlarides,L.L., Kineticsof theFischer-Tropschsynthesis,Ind.Eng.

Chem.ProcessDes.Dev. 1983, 22, 1–9.[14] Sarup,B.; Wojciechowski, B.W., Studiesof theFischer-Tropschsynthesison a

cobaltcatalyst.II. Kinetics of carbonmonoxideconversionto methaneandtohigherhydrocarbons,Can.J. Chem.Eng. 1989, 67, 62–74.

[15] Shen,W.J.;Zhou,J.L.; Zhang,B.J.,Kineticsof Fischer-Tropschsynthesisoverprecipitatediron catalyst,J. Nat.GasChem.1994, 4, 385–400.

[16] Zhang, H.-B.; Schrader, G.L., Characterisationof a fused iron catalyst forFischer-Tropschsynthesisby in situ laserRamanspectroscopy, J. Catal. 1985,95, 325–332.

[17] Lox, E.S.;Marin, G.B.; de Graeve, E.; Bussiere,P., Characterizationof a pro-motedprecipitatediron catalystfor Fischer-Tropschsynthesis,Appl. Catal. A

1988, 40, 197–218.[18] Shroff, M.D.; Kalakkad,D.S.; Coulter, K.E.; Kohler, S.D; Harrington,M.S.;

Jackson,N.B.; Sault,A.G.; Datye,A.K., Activationof iron precipitatedFischer-Tropschcatalysts,J. Catal.1995, 156, 185–207.

[19] Rao,K.R.P.M.; Huggins,F.E.; Mahajan,V.; Huffman,G.P.; Rao,V.U.S.;Bhatt,B.L.; Bukur, D.B.; Davis, B.H.; O’Brien, R.J.,Mossbauerspectroscopy studyof iron-basedcatalystsusedin Fischer-Tropschsynthesis,Top. Catal. 1995, 2,71–78.

[20] Rethwisch,D.G.; Dumesic,J.A., Adsorptive and catalytic propertiesof sup-portedmetaloxides.III. Water-gasshift over supportediron and zinc oxides,J. Catal.1986, 101, 35–42.

[21] Ponec,V., Someaspectsof the methanationand Fischer-Tropschsynthesis,Catal.Rev.-Sci.Eng. 1978, 18, 151–171.

[22] Anderson,R.B., Catalystsfor the Fischer-Tropsch synthesis, vol. 4, Van Nos-trandReinhold,New York 1956.

[23] Anderson,R.B., The Fischer-Tropsch synthesis, AcademicPress,New York

INTRINSIC K INETICS OF THE GAS-SOLID FT AND WGS REACTIONS 167

1984.[24] Bell, A.T., Catalyticsynthesisof hydrocarbonsover groupVIII metals.A dis-

cussionon thereactionmechanism,Catal.Rev.-Sci.Eng. 1981, 23, 203–232.[25] Hindermann,J.P.; Hutchings,G.J.;Kiennemann,A., Mechanisticaspectsof the

formationof hydrocarbonsandalcoholsfromCOhydrogenation,Catal.Rev.-Sci.

Eng. 1993, 35, 1–127.[26] Dry, M.E., Conversionof syngasto fuelsandchemicals,in Int. Conf. on Catal.

& Catal.Proc., CapeTown, SouthAfrica, 1993pp.57–66.[27] Dry, M.E., Practicalandtheoreticalaspectsof thecatalyticFischer-Tropschpro-

cess,Appl.Catal.A 1996, 138, 319–344.[28] Adesina,A.A., Hydrocarbonsynthesisvia Fischer-Tropschreaction:travailsand

triumphs,Appl.Catal.A 1996, 138, 345–367.[29] Fischer, F.; Tropsch,H., Uber die Herstellungsynthetischerolgemische(Syn-

thol) durchAufbau ausKohlenoxydund Wasserstoff, Brennst.Chem.1923, 4,276–285.

[30] Schulz,H.; VanSteen,E.; Claeys,M., Olefin formation,hydrogenationandiso-merizationin the kinetic regime of Fischer-Tropschsynthesis,in Selectivehy-

drogenationanddehydrogenation, DGMK, Kassel,Germany, 1993.[31] Bianchi,D.; Tau,L.M.; Borcar, S.;Bennett,C.O.,Natureof thespecieson sup-

portediron duringCO/H2 reaction,J. Catal.1983, 84, 358–374.[32] Ponec,V.; VanBarneveld, W.A., The role of chemisorptionin Fischer-Tropsch

synthesis,Ind. Eng. Chem.Prod.Res.Des.1979, 4, 268–271.[33] Kellner, C.S.;Bell, A.T., Thekineticsandmechanismof carbonmonoxidehy-

drogenationoveralumina-supportedruthenium,J. Catal.1981, 70, 418–432.[34] Hovi, J.-P.; Lahtinen,J.; Liu, Z.S.; Nieminen,R.M., MonteCarlostudyof CO

hydrogenationoncobaltmodelcatalysts,J. Chem.Phys.1995, 102, 7674–7682.[35] Graaf,G.H.;Winkelman,J.G.M.;Stamhuis,E.J.;Beenackers,A.A.C.M., Kinet-

icsof thethree-phasemethanolsynthesis,Chem.Eng. Sci.1988, 43, 2161–2168.[36] Froment,G.F.; Hosten,L.H., Catalytickinetics: modelling, in J.R.Anderson;

M. Boudart,eds.,Catalysis.Scienceand technology, vol. 2, Springer-Verlag,New York, 1981pp.98–170.

[37] Biloen,P.; Sachtler, W.M.H., Mechanismof hydrocarbonsynthesisoverFischer-Tropschcatalysts,Adv. Catal.1981, 30, 165–216.

[38] Dry, M.E.; Shingles,T.; Boshoff, L.J.; Oosthuizen,G.J., Heatsof adsorptionon promotediron surfacesandtherole of alkali in Fischer-Tropschsynthesis,J.

Catal.1969, 15, 190–199.

168 CHAPTER 5

[39] Yates,I.C.; Satterfield,C.N., Effect of carbondioxide on the kinetics of theFischer-Tropschsynthesison iron catalysts,Ind. Eng. Chem.Res.1989, 28, 9–12.

[40] Yates,I.C.; Satterfield,C.N., Intrinsic kineticsof theFischer-Tropschsynthesisonacobaltcatalyst,EnergyFuels1991, 5, 168–173.

[41] Dry, M.E.; Shingles,T.; Boshoff, L.J.; Oosthuizen,G.J.,Rateof the Fischer-Tropschreactionover iron catalysts,J. Catal.1972, 25, 99–104.

[42] Whiters, Jr., H.P.; Eleizer, K.F.; Mitchell, J.W., Slurry-phaseFischer-Tropschsynthesisandkinetic studiesover supportedcobaltcarbonyl derived catalysts,Ind. Eng. Chem.Res.1990, 29, 1807–1814.

[43] Deckwer, W.-D.; Kokuun,R.; Sanders,E.; Ledakowicz, S., Kinetic studiesofFischer-Tropschsynthesison suspendedFe/K catalyst.Rateinhibition by CO2

andH2O, Ind. Eng. Chem.ProcessDes.Dev. 1986, 25, 643–649.[44] Chanenchuk,C.A.; Yates,I.C.; Satterfield,C.N.,TheFischer-Tropschsynthesis

with amechanicalmixtureof acobaltcatalystandacopper-basedwatergasshiftcatalyst,EnergyFuels1991, 5, 847–855.

[45] Van der Laan, G.P.; Beenackers, A.A.C.M., Kinetics and selectivity of theFischer-Tropschsynthesis.A literaturereview, Acceptedfor publicationin Catal.

Rev.-Sci.Eng.

[46] Graaf,G.H.;Sijtsema,P.J.J.M.;Stamhuis,E.J.;Joosten,G.E.H.,Chemicalequi-libria in methanolsynthesis,Chem.Eng. Sci.1986, 41, 2883–2890.

[47] Bukur, D.B.; Nowicki, L.; Manne,R.K.; Lang, X., Activation studieswith aprecipitatediron catalystfor Fischer-Tropschsynthesis2. Reactionstudies,J.

Catal.1995, 155, 366–375.[48] Westerterp,K.R.; vanSwaaij,W.P.M.; Beenackers,A.A.C.M., Chemicalreactor

designandoperation, Wiley & Sons,secondedn.1984.[49] Jonker, G.H.; Veldsink, J.-W.; Beenackers, A.A.C.M., Intrinsic kinetics of 9-

monoenicfattyacidmethylesterhydrogenationovernickel-basedcatalysts,Ind.

Eng. Chem.Res.1997, 36, 1567–1579.[50] Graaf, G.H., The synthesisof methanolin gas-solidand gas-slurryreactors,

Ph.D.thesis,Universityof Groningen,TheNetherlands1988.[51] Fisher, R.A, Statisticalmethodsfor research workers, Macmillan, 14th edn.

1970.